Microsoft Accepts Ten AI Principles for Schools with AFT

Microsoft and the American Federation of Teachers agree on ten principles of safety and privacy for AI in schools. Districts will be able to integrate them into their contracts starting in November, against the backdrop of one-year bans in New York and Los Angeles and calls for firmness from the AFT president.

The AFT pushes for strict rules and funds skill development

Randi Weingarten considers the terms reached with Microsoft to be "foolproof" and believes they are necessary, as she claims no other party, including the federal government, has taken the initiative. She asserts that only legally binding provisions prevent remaining at the wish list stage. In May, the AFT president called for a ban on screens before third grade and for prohibiting AI intended for students before middle school. Additionally, the AFT announced the creation of a $23 million AI training center for educators, funded by Anthropic, Microsoft, and OpenAI. The union has been critical of technology in schools while remaining open to collaboration with tech companies to anticipate the arrival of AI.

What Microsoft agrees to and what districts can demand

Microsoft has accepted a set of safety and privacy principles for AI in schools as part of an agreement with the American Federation of Teachers (AFT), the second-largest teachers' union in the United States, and its New York affiliate, the United Federation of Teachers (UFT). The company commits to ten principles that school districts can contractually apply if they wish. Among these commitments are the prohibition of training AI models on student or educator data, limiting data collection, the obligation to inform families about how tools work in simple terms, banning AI companions, and requiring human review for high-risk decisions. Starting in November, school districts will be able to add these clauses to new or existing agreements without renegotiating their entire contracts.

Bans in New York and Los Angeles frame the period

This agreement follows the announcement, a week earlier, by two major school networks of a ban on AI tools for students. The cities of New York and Los Angeles have implemented one-year bans on these tools for a large number of students, allowing districts time to determine appropriate use and necessary protective measures.